Dr. Ben Sharp presented this invited paper during the plenary session at the Internoise 99 convention in Fort Lauderdale, Florida.
Introduction
The approach of the new millenium presents us with an ideal opportunity to pause and reflect on progress achieved in the environmental noise arena. This historical moment is a convenient time to assess the role that noise control has played in improving our quality of life, evaluate whether our progress has achieved stated goals, and predict what might be expected in the future. It is a time of reckoning, an opportunity for innovative thinking.
Forty years ago most major noise complaints were directed at jet aircraft, noisy trucks and construction equipment, and poor sound isolation in buildings and offices. Today, the complaints are associated with jet aircraft, highway and construction noise, and poor acoustics in buildings. An outsider might be excused for wondering whether we have made any progress at all. Has our quality of life improved? In a recent Editorial in Noise News (June 1999), the 1999 INCE President Stephen Roth noted that noise in public places is still unacceptable, even though technical solutions are available. The real answer is that we have made significant progress in some areas, but that overall, we have certainly not achieved the gains hoped for in the environmental decade of the 1970's. Technology that has been developed has not found its way into practice. And an increase in public sensitivity to transportation noise has tended to obscure some of the real gains made.
So, where do we really stand in our fight against noise, and what might we expect in the future? One way to answer this question is to examine a few representative examples in different areas of noise control as applied to building acoustics, machinery noise, and transportation noise. The attempt will be to assess progress and identify the major holes that are preventing us from making major advances in our quality of life.
Building Acoustics
The materials and design procedures necessary to satisfy the minimum standards required of most buildings are well known. However, there are many areas that need further research and development. Additional knowledge is required to increase the accuracy and utility of measurement and prediction procedures, so that the cost of noise control can be reduced. The techniques for predicting and mitigating noise generated by HVAC systems are based on decades-old data. Research in these and other areas has declined markedly in recent years, with the result that the techniques in use today are no different than those of four decades ago.
Great efforts have been made to ensure repeatability and reproducibility in test procedures. In the measurement of sound transmission loss, for example, much effort has been expended over the last forty years to develop a standard procedure. The result has been a requirement for large and expensive rooms in an attempt to obtain the diffuse sound fields necessary for true statistical testing. One might think that these efforts would have resulted in an acceptable test procedure, but results have shown otherwise. A "round robin" test recently conducted in 21 European laboratories on two fairly simple wall constructions showed variations of up to 10dB at low frequencies, and of 5dB at medium and high frequencies – quite a discouraging result in view of all the work that has gone into development of the test procedure. The problem goes far deeper than that of academic interest. A product manufacturer can sometimes get the result he needs by going to the right test laboratory. Furthermore, since laboratory measurements are used to test the validity of proposed theories, or to empirically adjust approximate theories, it is possible to conceive of theories that are facility dependent.
Unfortunately, with the current emphasis in the United States (California being an exception) on "type" specifications rather than "performance" standards, there is little incentive for designers to worry about the inadequacies of the test procedures. We need to introduce performance requirements in building design in order to ensure satisfactory interior noise environments. These need to be coupled with improved test procedures to eliminate overdesign and minimize construction costs.
The overlying problem here is that we still do not fully understand the influence of the sound fields in rooms on transmission and absorption measurements, particularly at low frequencies where many of today's noise problems exist. This lack of understanding becomes even more of a problem when the results are applied to actual buildings where the typical room sizes are much smaller than the test laboratory rooms.
There is always a reluctance on the part of Standards Committees to make radical changes to test procedures, but there are alternatives to the current procedures that can be considered. If the test rooms influence the results in ways we do not fully understand, why not develop a technique that does not require rooms? They are certainly very expensive to construct and maintain. Alternatively, use the improved understanding of sound fields (see above) to design rooms or appropriately normalize the results. Then, the true performance of structures and absorption materials could be obtained. This is probably the most important new technology that could change the way we design and construct buildings for noise control.
One of the other major problems in multi-family buildings is the prevalence of impact noise with today's lightweight floor structures. Improved design techniques are badly needed in this historically ignored area of building acoustics. Impact noise research has always lagged far behind that for airborne noise. The antiquated test procedure for impact noise measurement also needs to be changed, and this requires research into a better understanding and modeling of the process. We will not be able to reduce impact noise in dwellings until this research is complete.
We have known what needs to be done to prevent flanking noise and maximize the performance of structures in buildings. Every text book contains a list of simple techniques. The problem is that although simple, they can be extremely time-consuming to implement, and hence builders will ignore them unless they are mandatory. We need alternative approaches to building design that avoid flanking problems, coupled with performance type standards to ensure compliance.
The majority of effort related to building acoustics has been directed at intra-dwelling noise and providing protection from interior noise sources. However, in many locations, the main problems are from exterior noise sources, such as aircraft and highway vehicles. Generally, the same principles apply, but the type of structures are completely different. Research in this area has almost been totally ignored in the past despite the large amounts of government funds that have been directed at building retrofit by many countries to reduce this form of impact. The same problems in understanding the effects of room acoustics are present here as they are for interior structures.
Machinery Noise
Many of the same comments can be made for progress in machinery noise control as for building noise control, namely that, noise control has been generally confined to band-aid solutions using fairly familiar technology, such as enclosures, barriers, lagging, etc. Only limited effort has been applied to designing noise control into the product. Noise models are usually inaccurate in defining and quantifying all of the source mechanisms. We have vastly improved tools using numerical computation techniques, but they are still unable to model complete machines, just parts of them. Computation at all but the lowest frequencies is extremely computation intensive, with the result that statistical methods, such as SEA, are employed. And in these models, we still do not have a good handle on damping and transmission of energy through joints. These fundamentals must be understood before predictive tools can be used reliably for design purposes.
In the mid-80's active noise control was being touted as the answer to many of our noise control problems, specifically those in the low-frequency range. Fifteen years later, we are still waiting to reap the benefits of this new technology. To date, the only practical applications in noise control are in air conditioning ducts, some aircraft and automobile interiors, and noise canceling headphones. Other applications today are limited to high-value, limited production systems.
The problems in applying active noise control systems are in its current inability to work with complex sound fields, and the high cost of installations. Rugged systems for industry are not yet available. Presumably, some of these problems will be overcome with increases in the speed of DSP's. Others will require the development of sensing systems and actuators to control structural radiation, before the promises of active control will be achieved.
Transportation Noise
In the 1970's, the Environmental Protection Agency estimated that the major noise sources affecting our health and welfare were those related to transportation, particularly highway vehicles and aircraft. The regulations introduced in that decade have effectively eliminated heavy trucks with noisy mufflers and tires, and hence have reduced the peak noise levels alongside highways. But that does not mean that highway traffic is no longer a noise problem. The noise levels from individual automobiles have been reduced by up to 8dB in Europe as a result of strict regulations, but highway noise levels have dropped by only 1dB. One of the reasons is that traffic volumes have increased significantly over the same time period, negating the reduction in individual vehicle levels.
Automobile traffic still forms the background noise levels in most suburban areas, due to the noise from the tire/roadway interaction - commonly known as tire noise. The noise sources on automobiles have been quietened to the point where tire noise is dominant. Further reductions in these other sources are therefore not a major priority until tire noise levels are reduced. Research in this area in the United States has been fragmented, but is being driven now by regulations in Europe. Still at issue are the fundamental mechanisms of tire noise generation, and methods of incorporating quiet designs into a tire that also maintains durability and traction. At the other side of the interface, researchers in Europe have developed quiet road surface designs that reduce tire levels by 10dB or so. This technology needs to be studied and demonstrated in the United States. The modeling of tire noise mechanisms is the most important activity in the reduction of highway ambient levels.
The other main reason why highway noise levels have not dropped in accordance with vehicle noise levels is that the standard full-throttle test procedure for measuring automobile noise is not at all representative of typical operation. This fact has been known for over 25 years, and although various sporadic attempts have been made to modify the procedure, it is only recently that any progress has been made in the development of a multi-modal test that does represent typical operation. Development of such a procedure is paramount to a successful abatement policy.
One of the most spectacular examples of noise control, and an excellent example of noise control benefiting from improved performance, has been the reduction in jet engine noise. Average aircraft noise levels have been reduced by 20 to 25dB over the last 40 years, largely through the development by industry of high by-pass ratio jet engines, but also from improved acoustic nacelle liners. This has allowed gradual reductions in aircraft noise levels hastened by national phase-out regulations, but like highway noise, increases in numbers is negating reductions in airport noise. The easy work may have been accomplished, however, and further reductions are going to require considerable effort in controlling engine fan and core noise. As is the case with machinery noise in general, current models do not account satisfactorily for all source mechanisms. Furthermore, there is not a good relationship between the results obtained using computer codes and scale-model tests with those from installed engine tests.
Conclusions
One of the purposes of this article was to assess the current state-of-the-art in noise control. In discussions with workers in the field, an almost unanimous consensus is that many gaps exist in our understanding of fundamental mechanisms. Over the past 40 years we have achieved noise control largely through band-aid measures, without getting to the heart of the problem, namely, the noise source itself. In other words, we have picked the low-lying fruit.
In specific terms, it is necessary to address fundamentals in building acoustics and machinery noise. Equally importantly, we must revisit and rewrite the test procedures in many areas so that they are representative of typical operation and rank products appropriately.
But there is also another problem with our noise control efforts, and this is the institutional problem noted by Stephen Roth. Much of what we have learned has never been implemented for a number of reasons, cost and demand being two of the most important. Unless there is a significant cost or marketing advantage, or regulation, manufacturers generally will not modify their products to reduce noise. It is therefore up to us to concentrate more on developing affordable noise control solutions, and in working with legislators to enact appropriate regulations to ensure that any new technology we develop is applied.
-End-
By Dr. Ben H. Sharp
12/99
Dr. Ben Sharp is the Director of the Acoustics Group of Wyle in Arlington, VA, USA. He can be reached at (703) 415-4550 Ext. 15, or by e-mail at bsharp@arl.wyle.com
environment@wyle.com